SHACHI - Language Resource Metadata Database

The orthographic standard differs according to periods and geographical regions. This database selects the manuscripts of Buddhist and historical texts and records various examples of Chinese glyphs.

The database collects the usages of Chinese readings and morphemes based on the analysis of the documents such as Rime dictionaries or Phonetic Glosses of Dharani between 15-18 centuries in Korea. The image samples from the original texts are partially available.

Among 5,300,000 blogs at 28 domestic companies of blog business, 600,000,000 articles have been retrieved since January, 2007.

Large-scale parallel corpus of Japanese-English corresponding data from American patent collection and Japanese patent collection.

Sound database of 100 names of places. 12 males each pronounced the words twice.

Novels, essays, biographies, political criticisms / white paper, law-affiliated documents / treaty documents, poetry, are collected and made into Chinese-Japanese Bilingual Corpus.

It is bundled with digital video and audio data (movies) as well as full texts. The unique features of our hypermedia corpus will be as follows:
1. the intonation, pause and some other information can be represented in their original forms by digital sounds, thereby making it unnecessary to assign specific linguistic notations or 'esoteric' symbols
2. non-verbal information is available in the form of multi-angle digital movies, providing the speaker/listener's facial expressions, noddings, and so on
3. any movie and sound data is randomly accessible by respective code number and/or one or more key words in addition to the traditional text search thanks to new digital video-on-demand (VOD) technology and hypertext description languages such as HTML/XML.

The feature 1 will enable novice users of language databases, especially language teachers, to use them as their teaching materials since they are not required to learn anything about linguistic notations and other complicated assumptions. The features 2 and 3 will extend the range of conversation analysis to deal with extra-linguistic factors like gestures and even atmospheric conditions that have virtually been excluded from the subjects of studies.